TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

SotA/Natural Language Processing/Question Answering/SQuAD1.1

Question Answering on SQuAD1.1

Metric: EM (higher is better)

LeaderboardDataset
Loading chart...

Results

Submit a result
#Model↕EM▼Extra DataPaperDate↕Code
1{ANNA} (single model)90.622Yes---
2LUKE (single model)90.202YesLUKE: Deep Contextualized Entity Representations...2020-10-02Code
3LUKE (single model)90.202NoLUKE: Deep Contextualized Entity Representations...2020-10-02Code
4LUKE90.2NoLUKE: Deep Contextualized Entity Representations...2020-10-02Code
5XLNet (single model)89.898NoXLNet: Generalized Autoregressive Pretraining fo...2019-06-19Code
6XLNet (single model)89.898YesXLNet: Generalized Autoregressive Pretraining fo...2019-06-19Code
7XLNET-123++ (single model)89.856No---
8XLNET-123+ (single model)89.709No---
9XLNET-123 (single model)89.646No---
10Unnamed submission by NMC88.912No---
11BERTSP (single model)88.912No---
12SpanBERT (single model)88.839No---
13SpanBERT (single model)88.8NoSpanBERT: Improving Pre-training by Representing...2019-07-24Code
14BERT+WWM+MT (single model)88.65No---
15Tuned BERT-1seq Large Cased (single model)87.465No---
16LinkBERT (large)87.45NoLinkBERT: Pretraining Language Models with Docum...2022-03-29Code
17BERT (ensemble)87.433NoBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
18BERT-LARGE (Ensemble+TriviaQA)87.4NoBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
19ATB (single model)86.94No---
20Tuned BERT Large Cased (single model)86.521No---
21BERT+MT (single model)86.458No---
22Knowledge-enhanced BERT (single model)85.944No---
23KT-NET (single model)85.944No---
24ST_bl85.43No---
25nlnet (ensemble)85.356No---
26EL-BERT (single model)85.335No---
27BISAN (single model)85.314No---
28BERT+Sparse-Transformer85.125No---
29BERT (single model)85.083YesBERT: Pre-training of Deep Bidirectional Transfo...2018-10-11Code
30DPN (single model)84.978No---
31BERT-uncased (single model)84.926No---
32WD (single model)84.402No---
33Original BERT Large Cased (single model)84.328No---
34MARS (ensemble)83.982No---
35Common-sense Governed BERT-123 (single model)83.93No---
36WD1 (single model)83.804No---
37nlnet (single model)83.468No---
38Pytalk + Stanza + BERT (single model)83.426No---
39Reinforced Mnemonic Reader + A2D (ensemble model)82.849No---
40BERT-Base mod (single model)82.681No---
41r-net+ (ensemble)82.65No---
42Hybrid AoA Reader (ensemble)82.482No---
43QANet (single)82.471No---
44SLQA+ (ensemble)82.44No---
45Reinforced Mnemonic Reader (ensemble model)82.283YesReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
46r-net (ensemble)82.136No---
47BERT (single model)82.062No---
48AttentionReader+ (ensemble)81.79No---
49MMIPN81.58No---
50BERT - 6 Layers81.5NoInformation Theoretic Representation Distillation2021-12-01Code
51KACTEIL-MRC(GF-Net+) (ensemble)81.496No---
52Reinforced Mnemonic Reader + A2D + DA (single model)81.401No---
53ARSG-BERT (single model)81.307No---
54BERT-COMPOUND-DSS (single model)81.045No---
55BiDAF + Self Attention + ELMo (ensemble)81.003NoDeep contextualized word representations2018-02-15Code
56BiDAF + Self Attention + ELMo (ensemble)81.003NoDeep contextualized word representations2018-02-15-
57BERT-COMPOUND (single model)80.72No---
58mBERT + Task Adapter (Single)80.667No---
59AVIQA+ (ensemble)80.615No---
60Reinforced Mnemonic Reader + A2D (single model)80.489No---
61SLQA+80.436No---
62{EAZI} (ensemble)80.436No---
63EAZI+ (ensemble)80.426No---
64DNET (ensemble)80.164No---
65Hybrid AoA Reader (single model)80.027No---
66BiDAF + Self Attention + ELMo + A2D (single model)79.996No---
67r-net+ (single model)79.901No---
68batch (single model)79.859No---
69MAMCN+ (single model)79.692No---
70MAMCN+ (single model)79.692No---
71SAN (ensemble model)79.608NoStochastic Answer Networks for Machine Reading C...2017-12-10Code
72BERT-INDEPENDENT-DSS-FILTERED (single model)79.597No---
73Reinforced Mnemonic Reader (single model)79.545NoReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
74SLQA+ (single model)79.199No---
75Interactive AoA Reader+ (ensemble)79.083No---
76MIR-MRC(F-Net) (single model)79.083No---
77KACTEIL-MRC(GF-Net+Distillation) (single model)79.083No---
78KACTEIL-MRC (GF-Net+Distillation)79.083No---
79MDReader79.031No---
80FusionNet (ensemble)78.978NoFusionNet: Fusing via Fully-Aware Attention with...2017-11-16Code
81DCN+ (ensemble)78.852NoDCN+: Mixed Objective and Deep Residual Coattent...2017-10-31Code
82KACTEIL-MRC(GF-Net+) (single model)78.664No---
83KACTEIL-MRC (GF-Net+)78.664No---
84BERT-INDEPENDENT (single model)78.653No---
85BiDAF + Self Attention + ELMo (single model)78.58NoDeep contextualized word representations2018-02-15Code
86BiDAF + Self Attention + ELMo (single model)78.58NoDeep contextualized word representations2018-02-15-
87aviqa (ensemble)78.496No---
88KakaoNet (single model)78.401No---
89SLQA(ensemble)78.328No---
90SLQA (ensemble)78.328No---
91MEMEN (single model)78.234NoMEMEN: Multi-layer Embedding with Memory Network...2017-07-28-
92MEMEN (single model)78.234NoMEMEN: Multi-layer Embedding with Memory Network...2017-07-28-
93BiDAF++ with pair2vec (single model)78.223No---
94MDReader078.171No---
95test78.087No---
96Interactive AoA Reader (ensemble)77.845No---
97BERT - 3 Layers77.7NoInformation Theoretic Representation Distillation2021-12-01Code
98DNET (single model)77.646No---
99RaSoR + TR + LM (single model)77.583NoContextualized Word Representations for Reading ...2017-12-10Code
100BiDAF++ (single model)77.573No---
101AttentionReader+ (single)77.342No---
102Jenga (ensemble)77.237No---
103{gqa} (single model)77.09No---
104Conductor-net (ensemble)76.996NoPhase Conductor on Multi-layered Attentions for ...2017-10-28-
105MARS (single model)76.859No---
106SAN (single model)76.828NoStochastic Answer Networks for Machine Reading C...2017-12-10Code
107VS^3-NET (single model)76.775No---
108r-net (single model)76.461No---
109r-net (single model)76.461No---
110FRC (single model)76.24No---
111QANet + data augmentation ×376.2NoQANet: Combining Local Convolution with Global S...2018-04-23Code
112Conductor-net (ensemble)76.146No---
113KAR (single model)76.125NoExplicit Utilization of General Knowledge in Mac...2018-09-10-
114smarnet (ensemble)75.989No---
115FusionNet (single model)75.968NoFusionNet: Fusing via Fully-Aware Attention with...2017-11-16Code
116AVIQA-v2 (single model)75.926No---
117Interactive AoA Reader+ (single model)75.821No---
118RaSoR + TR (single model)75.789NoContextualized Word Representations for Reading ...2017-12-10Code
119MEMEN (ensemble)75.37NoMEMEN: Multi-layer Embedding with Memory Network...2017-07-28-
120Mixed model (ensemble)75.265No---
121two-attention-self-attention (ensemble)75.223No---
122Kbs (single model)75.034No---
123ReasoNet (ensemble)75.034YesReasoNet: Learning to Stop Reading in Machine Co...2016-09-17-
124EfficientQA 125M74.9NoEfficientQA : a RoBERTa Based Phrase-Indexed Que...2021-01-06-
125DCN+ (single model)74.866NoDCN+: Mixed Objective and Deep Residual Coattent...2017-10-31Code
126eeAttNet (single model)74.604No---
127SLQA (single model)74.489No---
128Conductor-net (single model)74.405NoPhase Conductor on Multi-layered Attentions for ...2017-10-28-
129Mnemonic Reader (ensemble)74.268NoReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
130S^3-Net (ensemble)74.121No---
131SEDT (ensemble model)74.09NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
132SSAE (ensemble)74.08No---
133Multi-Perspective Matching (ensemble)73.765NoMulti-Perspective Context Matching for Machine C...2016-12-13Code
134BiDAF (ensemble)73.744NoBidirectional Attention Flow for Machine Compreh...2016-11-05Code
135SEDT+BiDAF (ensemble)73.723NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
136Interactive AoA Reader (single model)73.639No---
137Jenga (single model)73.303No---
138Conductor-net (single)73.24NoPhase Conductor on Multi-layered Attentions for ...2017-10-28-
139jNet (ensemble)73.01NoExploring Question Understanding and Adaptation ...2017-03-14-
140T-gating (ensemble)72.758No---
141two-attention-self-attention (single model)72.6No---
142Conductor-net (single)72.59No---
143AVIQA (single model)72.485No---
144BiDAF + Self Attention (single model)72.139NoSimple and Effective Multi-Paragraph Reading Com...2017-10-29Code
145S^3-Net (single model)71.908No---
146QFASE71.898No---
147attention+self-attention (single model)71.698No---
148Dynamic Coattention Networks (ensemble)71.625NoDynamic Coattention Networks For Question Answer...2016-11-05Code
149smarnet (single model)71.415NoSmarnet: Teaching Machines to Read and Comprehen...2017-10-08-
150SRU71.4NoSimple Recurrent Units for Highly Parallelizable...2017-09-08Code
151AttReader (single)71.373No---
152DCN + Char + CoVe71.3NoLearned in Translation: Contextualized Word Vect...2017-08-01Code
153M-NET (single)71.016No---
154Mnemonic Reader (single model)70.995NoReinforced Mnemonic Reader for Machine Reading C...2017-05-08Code
155MAMCN (single model)70.985No---
156FastQAExt70.849NoMaking Neural QA as Simple as Possible but not S...2017-03-14Code
157RaSoR (single model)70.849NoLearning Recurrent Span Representations for Extr...2016-11-04Code
158Document Reader (single model)70.733NoReading Wikipedia to Answer Open-Domain Questions2017-03-31Code
159Ruminating Reader (single model)70.639NoRuminating Reader: Reasoning with Gated Multi-Ho...2017-04-24-
160jNet (single model)70.607NoExploring Question Understanding and Adaptation ...2017-03-14-
161ReasoNet (single model)70.555NoReasoNet: Learning to Stop Reading in Machine Co...2016-09-17-
162Multi-Perspective Matching (single model)70.387NoMulti-Perspective Context Matching for Machine C...2016-12-13Code
163DrQA70NoReading Wikipedia to Answer Open-Domain Questions2017-03-31Code
164SimpleBaseline (single model)69.6No---
165SSR-BiDAF69.443No---
166SEDT+BiDAF (single model)68.478NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
167FastQA68.436NoMaking Neural QA as Simple as Possible but not S...2017-03-14Code
168PQMN (single model)68.331No---
169SEDT (single model)68.163NoStructural Embedding of Syntactic Trees for Mach...2017-03-02-
170T-gating (single model)68.132No---
171BiDAF (single model)67.974NoBidirectional Attention Flow for Machine Compreh...2016-11-05Code
172Match-LSTM with Ans-Ptr (Boundary) (ensemble)67.901NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
173FABIR67.744NoA Fully Attention-Based Information Retriever2018-10-22Code
174AllenNLP BiDAF (single model)67.618No---
175BIDAF-COMPOUND-DSS (single model)67.544No---
176Iterative Co-attention Network67.502No---
177newtest66.527No---
178BIDAF-INDEPENDENT-DSS (single model)66.516No---
179Dynamic Coattention Networks (single model)66.233NoDynamic Coattention Networks For Question Answer...2016-11-05Code
180DCN66.2NoDynamic Coattention Networks For Question Answer...2016-11-05Code
181MPCM65.5NoMulti-Perspective Context Matching for Machine C...2016-12-13Code
182BIDAF-COMPOUND (single model)65.163No---
183BIDAF-INDEPENDENT (single model)64.932No---
184Match-LSTM with Bi-Ans-Ptr (Boundary)64.744NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
185Unnamed submission by ravioncodalab64.439No---
186OTF dict+spelling (single)64.083NoLearning to Compute Word Embeddings On the Fly2017-06-01-
187Attentive CNN context with LSTM63.306No---
188OTF spelling (single)62.897NoLearning to Compute Word Embeddings On the Fly2017-06-01-
189OTF spelling+lemma (single)62.604NoLearning to Compute Word Embeddings On the Fly2017-06-01-
190Dynamic Chunk Reader62.499NoEnd-to-End Answer Chunk Extraction and Ranking f...2016-10-31-
191Fine-Grained Gating62.446NoWords or Characters? Fine-grained Gating for Rea...2016-11-06Code
192RQA+IDR (single model)61.145NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
193RQA+IDR (single model)61.145NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
194Match-LSTM with Ans-Ptr (Boundary)60.474NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
195Unnamed submission by Will_Wu59.058No---
196RQA (single model)55.827NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
197RQA (single model)55.827NoHarvesting and Refining Question-Answer Pairs fo...2020-05-06Code
198Match-LSTM with Ans-Ptr (Sentence)54.505NoMachine Comprehension Using Match-LSTM and Answe...2016-08-29Code
199UQA (single model)53.698No---
200Unnamed submission by jinhyuklee52.544No---
201Unnamed submission by minjoon52.533No---
202UnsupervisedQA V1 (ensemble)47.341No---
203UnsupervisedQA V1 (single model)44.215No---
204QANet (single model)12.273No---
2050No---
206QANet (ensemble)0No---
207superman-new-des0No---
208WAHnGREA0No---
209superman-des0No---
210XLNet-deep (ensemble)0No---